Field of the Invention
This invention generally relates to communication systems. More particularly, an exemplary embodiment of this invention relates to anomaly detection in communications systems.
Description of Related Art
Cyclic Redundancy Checksum (CRC) error detection is a common method of detecting errors in a data stream transmitted over a communications channel. ITU standard G.992.3, which is incorporated herein by reference in its entirety, describes CRC operations for ADSL systems in section 7.7.1.2. As discussed in G.992.3, the transmitter computes the transmitter CRC bits based on the transmitted bit stream and sends the CRC bits to the receiver. The receiver also computes the CRC bits based on the received bit stream and compares the locally computed CRC bits to the received CRC bits that were sent from the transmitter. If the receiver and transmitter CRC bits are identical, then the CRC computation indicates that there are no errors in the received bit stream. If however the received and transmitted CRC bits are not identical, then the CRC computation indicates that there are errors in the received bit stream.
DSL systems, and communications systems in general, use CRC errors, which are also known as anomalies, to diagnose and detect problematic service conditions. These CRC anomalies are typically computed, counted and reported based on some fundamental assumptions on how often the CRCs are computed. For example, in an ADSL systems, such as those specified in G.992.3, Severely Errored Seconds (SESs) are defined as 18 or more CRC anomalies in a 1-second interval. This corresponds to approximately 30 percent of computed CRCs being in error if the CRC is computed every 17 ms. The G.992.3 ADSL standard requires that the CRC is computed every 15 to 20 msecs. In ADSL 2 and VDSL 2 systems, the period of the CRC computation is called the period of the overhead channel (PERp). The G.992.3 standard requires that:15 ms≤PERp≤20 ms.